When nodejs is build, a qemu wrapper script is used to execute some
programs built for the target in user-mode emulation. However, when the
target and build machines are similar (e.g. x86_74), running those
programs fails, with errors such as:
cd ../../tools/v8_gypfiles; python ../../deps/v8/tools/run.py ../../out/Release/v8-qemu-wrapper ../../out/Release/bytecode_builtins_list_generator ../../out/Release/obj.host/gen/generate-bytecode-output-root/builtins-generated/bytecodes-builtins-list.h
../../out/Release/bytecode_builtins_list_generator: /lib/x86_64-linux-gnu/libstdc++.so.6: version `GLIBCXX_3.4.29' not found (required by ../../out/Release/bytecode_builtins_list_generator)
../../out/Release/bytecode_builtins_list_generator: /lib/x86_64-linux-gnu/libc.so.6: version `GLIBC_2.33' not found (required by ../../out/Release/bytecode_builtins_list_generator)
../../out/Release/bytecode_builtins_list_generator: /lib/x86_64-linux-gnu/libc.so.6: version `GLIBC_2.32' not found (required by ../../out/Release/bytecode_builtins_list_generator)
../../out/Release/bytecode_builtins_list_generator: /lib/x86_64-linux-gnu/libc.so.6: version `GLIBC_2.34' not found (required by ../../out/Release/bytecode_builtins_list_generator)
Return code is 1
So the question is: why the heck does Qemu use the host C library?
To answer this question, we first have to look at how the -L option of
Qemu is implemented. This option is documented as such:
-L path QEMU_LD_PREFIX set the elf interpreter prefix to 'path'
The v8-qemu-wrapper script makes this option point to $(STAGING_DIR),
so that the ELF interpreter used is the one in $(STAGING_DIR).
However, contrary to what the option documentation says, this option
does much more than setting the ELF interpreter prefix: it is going to
affect how *all* system calls manipulating files (open, etc.) are
going to work.
When this option is passed, the function init_paths() in
https://git.qemu.org/?p=qemu.git;a=blob;f=util/path.c is called at
initialization time, and essentially its sets the global "base"
variable to point to the directory passed as -L argument.
Then, for every single syscall that manipulates a path, this path will
be passed through the path() function in the same file. This function
will first attempt to resolve the path with "base" as a prefix, and if
not, return the unprefixed path.
After adding some traces into this function, I was able to understand
what happens:
(1) -L$(STAGING_DIR) is passed, causing "base" to point to
$(STAGING_DIR)
(2) The target ELF interpreter from $(STAGING_DIR) is properly invoked
(3) When this ELF interpreter then resolves the libc.so.6 library, it
first looks for /etc/ld.so.cache.
(4) Qemu first looks for /etc/ld.so.cache with the -L prefix, i.e
$(STAGING_DIR)/etc/ld.so.cache, but it does not exist. So, the Qemu
system call emulation falls back to /etc/ld.so.cache, which means
the target ELF interpreter reads the /etc/ld.so.cache of the host
system.
(5) This /etc/ld.so.cache of the host system says that libc.so.6 is in
/lib/x86_64-linux-gnu/
(6) The target ELF interpreter therefore tries to use
/lib/x86_64-linux-gnu/libc.so.6. The Qemu system call emulation
first tries $(STAGING_DIR)/lib/x86_64-linux-gnu/libc.so.6, but
this library does not exist (it is in
$(STAGING_DIR)/lib/libc.so.6), so the Qemu system call emulation
falls back to /lib/x86_64-linux-gnu/libc.so.6 of the host system,
which exist... but is too old compared to the target C library.
Indeed, results from ld.so.cache take precedence over the simple
resolution of library paths in /usr/lib and /lib.
We see 3 possible ideas to resolve this problem:
(A) Change the behavior of Qemu to not fallback to unprefixed paths:
when -L is passed, all path-related system calls should see the
paths prefixed by the -L option.
Issue with this is that this change is unlikely to get accepted by
Qemu upstream. And there might be some side effects we have not
really identified.
(B) Create an empty $(STAGING_DIR)/etc/ld.so.cache. We have tested
this solution and it works: it gets used instead of the host
/etc/ld.so.cache. Because $(STAGING_DIR)/etc/ld.so.cache is empty,
there's no libc.so.6 match, so the target ELF interpreter goes
through its normal library location resolution logic, which falls
back to trying in /usr/lib and /lib, which works as those paths
ends up being prefixed with $(STAGING_DIR) by Qemu.
(C) Pass LD_LIBRARY_PATH pointing to $(STAGING_DIR)/lib and
$(STAGING_DIR)/usr/lib in the Qemu wrapper. This works because
LD_LIBRARY_PATH paths have precedence over paths given by
ld.so.cache.
This is the solution already used by the GOI qemu wrapper in
package/gobject-introspection/g-ir-scanner-qemuwrapper.in.
We chose to go with the third option, because it has been proven to work
for the GOI wrapper, and has been reported to solve #14366. Even though
the first option would be the best, it is also the one that has the
least chances to land any time soon (if ever); the second has not been
exercised, and the impact is not fully understood either (e.g what about
non-glibc toolchains?).
Fixes: #14366
Signed-off-by: Jens Maus <mail@jens-maus.de>
[yann.morin.1998@free.fr:
- add whole analsys done by Thomas in:
https://lore.kernel.org/buildroot/20221031213926.50d3c778@windsurf/
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 278d1db56b)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Unless told otherwise, ninja will spawn as many jobs as there are CPU
(plus 2). Nodejs is built with ninja, but it is a generic package, so
there is no variable (like with cmake-package) that passes the proper
number of parallel jobs as configured by the user.
As a consequence, the nodejs build will use as many CPU as are
available, possibly overcommitting the rsources the user expected to be
used.
Set the JOBS variableto limit that number.
Signed-off-by: Jens Maus <mail@jens-maus.de>
[yann.morin.1998@free.fr: reword commit log]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 84c24ab1b5)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This commit fixes a problem where the NUT package couldn't be
used as a NUT server due to the fact that the default group for
nobody is "nogroup" and not "nobody" like the internal default
of NUT. Thus, when starting a NUT server daemon the daemon starts
with incorrect group permissions. This commit fixes this
shortcoming by introducing a dedicated 'nut' user and 'nut' group
to drop priviledges to it.
Signed-off-by: Jens Maus <mail@jens-maus.de>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit cd46e1b143)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Sometimes it happens that a Company or a Physical Person sponsors the
creation and/or the upstreaming process of a patch, but at the moment
there is no way to give credits to it. In Linux they prepend '+sponsor'
to the e-mail of the contributor in both authorship and commit log tag as
discussed here[0]. So let's describe in the manual how to do that as a
standard.
[0]: https://lore.kernel.org/linux-doc/20230817220957.41582-1-giulio.benetti@benettiengineering.com/
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
[yann.morin.1998@free.fr:
- reword to reference sub-addressing and the RFC
- move to the "submitting patches" section, that already deals with
SoB tags
- differentiate between Your/Their names
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit de349df08c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The affected code isn't present in any release, see [0].
[0]: https://www.libssh.org/2023/07/14/cve-2023-3603-potential-null-dereference-in-libsshs-sftp-server/
The CPE entry for this CVE is
cpe:2.3:a:libssh:libssh:-:*:*:*:*:*:*:*
We interpret the "-" as matching any version. It actually means
"unspecified version", which is the cop-out in case there is nothing
useful to match. We can't really make our infrastructure ignore "-"
entirely, because for all we know our version is an unreleased commit
sha which _is_ vulnerable. Thus, the only way out is an exclusion which
we'll never be able to remove.
Signed-off-by: Daniel Lang <dalang@gmx.at>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
(cherry picked from commit a34a370f4e)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Per default, the fio package uses the "-march=native" GCC option. This
is of course wildly inappropriate for cross-compilation and can result
in illegal instructions. Thus we make sure fio will not use that
compiler option by adding --disable-native to FIO_OPTS.
Signed-off-by: Jens Maus <mail@jens-maus.de>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
(cherry picked from commit 056958724b)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The commit 4e365d1768 "package/tcl: bump to version 8.6.13" did NOT
refreshed the package patch, because the patch was still applying
correctly and the package was working as expected.
It was refreshed in the previous bump, in commit 9cf314745a
"package/tcl: bump to version 8.6.12". This was part of 2022.02.
Looking closer at the patch content, the -/+ lines are exactly the
same. So this patch does not change anything. Since the file was kept
and the commit log mention a patch refresh, the intent was more
likely to carry over the old patch (which was declaring all libc
functions as "unbroken".
This commit actually refreshes this patch. It was regenerated with
git format-patch. Since the patch is renamed due to git format-patch,
the .checkpackageignore is updated accordingly.
Note:
This ancient patch will be removed soon, as an upstream commit [1],
not yet in a release, cleaned up and removed those old parts.
[1] 04d66a2571
Cc: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
(cherry picked from commit ec8a9cc518)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As off 2022-11-22 CVE-2022-39377 is listed as affecting sysstat
< 2.16.1 instead of < 2.17.1. The text is not updated, but the CPE info
is.
Signed-off-by: Daniel Lang <dalang@gmx.at>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
(cherry picked from commit 6425e0b848)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The install to staging commands of the ne10 package are careful to
install the shared libraries only if they are built, but we forgot to
use the same care for the install to target commands, causing a build
failure on BR2_STATIC_LIBS=y configurations as no shared library was
built:
cp: cannot stat '/home/autobuild/autobuild/instance-15/output-1/build/ne10-1.2.1/modules/libNE10*.so*': No such file or directory
This commit fixes this by guarding the target installation commands to
BR2_STATIC_LIBS being empty.
The problem exists since the package was introduced in commit
318f3db0dc ("ne10: new package"), a good
10 years ago. Most likely it was not seen for many years as this
package is only available for ARM with NEON and AArch64, and we were
not testing fully static builds, except for ARMv5 that don't have
NEON. Now that we are doing more random testing, the problem started
being visible.
Fixes:
http://autobuild.buildroot.net/results/45b2c1af052271bc2f1bb96544f138d29e4f7dfd/
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
(cherry picked from commit 0b764a7d1e)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As off 2021-05-17 NVD added 1.19 as the first version that isn't
affected by CVE-2007-4476.
Signed-off-by: Daniel Lang <dalang@gmx.at>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 487c12a1f2)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
CVE-2022-45061 affects python <= 3.7.15, 3.8.0 through 3.8.15,
3.9.0 through 3.9.15, 3.10.0 through 3.10.8
The mentioned patch was removed in c38de813 when bumping to 3.11.1.
Signed-off-by: Daniel Lang <dalang@gmx.at>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 43dbfe4670)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
CVE-2021-32749 affects fail2ban <= 0.9.7, 0.10.0 through 0.10.6, and
0.11.0 through 0.11.2.
The mentioned patch was removed in 76853089 when bumping to 1.0.1.
Signed-off-by: Daniel Lang <dalang@gmx.at>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit a01a6b8dc8)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following build failures raised since bump to version 1.4.0 in
commit 456a739831:
-- Found PythonInterp: /usr/bin/python3.6 (found suitable version "3.6.9", minimum required is "3.6")
CMake Error at /home/buildroot/autobuild/run/instance-3/output-1/host/sparc64-buildroot-linux-gnu/sysroot/usr/share/cmake/pybind11/FindPythonLibsNew.cmake:147 (message):
Python config failure:
Traceback (most recent call last):
File "<string>", line 6, in <module>
ImportError: cannot import name 'sysconfig'
and
In file included from /home/buildroot/autobuild/instance-1/output-1/host/include/python3.11/Python.h:38,
from /home/buildroot/autobuild/instance-1/output-1/host/sh4-buildroot-linux-gnu/sysroot/usr/include/pybind11/detail/common.h:266,
from /home/buildroot/autobuild/instance-1/output-1/host/sh4-buildroot-linux-gnu/sysroot/usr/include/pybind11/attr.h:13,
from /home/buildroot/autobuild/instance-1/output-1/host/sh4-buildroot-linux-gnu/sysroot/usr/include/pybind11/detail/class.h:12,
from /home/buildroot/autobuild/instance-1/output-1/host/sh4-buildroot-linux-gnu/sysroot/usr/include/pybind11/pybind11.h:13,
from /home/buildroot/autobuild/instance-1/output-1/host/sh4-buildroot-linux-gnu/sysroot/usr/include/pybind11/numpy.h:12,
from /home/buildroot/autobuild/instance-1/output-1/build/zxing-cpp-2.1.0/wrappers/python/zxing.cpp:18:
/home/buildroot/autobuild/instance-1/output-1/host/include/python3.11/pyport.h:601:2: error: #error "LONG_BIT definition appears wrong for platform (bad gcc/glibc config?)."
601 | #error "LONG_BIT definition appears wrong for platform (bad gcc/glibc config?)."
| ^~~~~
Fixes:
- http://autobuild.buildroot.org/results/665b246a4bb14480152ee59050672a7469148a5b
- http://autobuild.buildroot.org/results/0502b05020de57e4910125c699c4264047187c51
- http://autobuild.buildroot.org/results/c5e7fe83d46c704e05800e3ae62bf476458c7b71
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 38f39a6031)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following build failure with oprofile raised since bump of
binutils to version 2.40 in commit
35656482d3:
configure: error: bfd library not found
[...]
configure:17928: checking for bfd_openr in -lbfd
configure:17953: /home/buildroot/autobuild/run/instance-1/output-1/host/bin/arm-linux-gcc -o conftest -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -Os -g0 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 conftest.c -lbfd -liberty -lpopt -ldl -lintl >&5
/home/buildroot/autobuild/run/instance-1/output-1/host/opt/ext-toolchain/bin/../lib/gcc/arm-buildroot-linux-uclibcgnueabi/12.3.0/../../../../arm-buildroot-linux-uclibcgnueabi/bin/ld: warning: libsframe.so.0, needed by /home/buildroot/autobuild/run/instance-1/output-1/host/arm-buildroot-linux-uclibcgnueabi/sysroot/usr/lib/libbfd.so, not found (try using -rpath or -rpath-link)
Indeed, in this case, libsframe is not installed even after applying
commit 1b4d921e1d because
BR2_BINUTILS_VERSION_2_40_X is not selected by anyone (binutils package
is selected by oprofile and the toolchain is not generated by buildroot)
To fix this issue, invert the logic: install libsframe by default (i.e.
when binutils is selected or with a buildroot toolchain). libsframe will
not be installed only if binutils < 2.40 is detected.
Fixes:
- http://autobuild.buildroot.org/results/af9a2d52823a332b48e6df14d2708b6a4b3833a4
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit e9f2f48a7e)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following build failure with gcc 4.8 raised since bump of snmppp
to version 3.5.0 in commit e011fa0415:
configure: error: Cannot find suitable libsnmp++ library
[...]
configure:9496: checking if libsnmp++ can be linked with flags from pkg-config
configure:9528: /home/buildroot/autobuild/run/instance-1/output-1/host/bin/arm-none-linux-gnueabi-g++ -o conftest -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -Os -g0 -D_FORTIFY_SOURCE=1 -pthread -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE -D_XOPEN_SOURCE=XPG6 conftest.cpp -L/home/buildroot/autobuild/run/instance-1/output-1/host/bin/../arm-buildroot-linux-gnueabi/sysroot/usr/lib -lsnmp++ >&5
In file included from /home/buildroot/autobuild/run/instance-1/output-1/host/arm-buildroot-linux-gnueabi/sysroot/usr/include/snmp_pp/snmp_pp.h:71:0,
from conftest.cpp:92:
/home/buildroot/autobuild/run/instance-1/output-1/host/arm-buildroot-linux-gnueabi/sysroot/usr/include/snmp_pp/uxsnmp.h:628:35: error: 'nullptr' was not declared in this scope
CSNMPMessage *snmp_message = nullptr);
^
Fixes:
- http://autobuild.buildroot.org/results/f272473e7b588f5390b183072935a0217290ee4e
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Tested-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Reviewed-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 7c6c018ad4)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- Drop patches (already in version) and so autoreconf
- Update COPYING hash (gpl mailing address updated with
9bd45cc06e6a5997fbd6)
- Fix CVE-2022-43634: This vulnerability allows remote attackers to
execute arbitrary code on affected installations of Netatalk.
Authentication is not required to exploit this vulnerability. The
specific flaw exists within the dsi_writeinit function. The issue
results from the lack of proper validation of the length of
user-supplied data prior to copying it to a fixed-length heap-based
buffer. An attacker can leverage this vulnerability to execute code in
the context of root. Was ZDI-CAN-17646.
- Fix CVE-2022-45188: Netatalk through 3.1.13 has an afp_getappl
heap-based buffer overflow resulting in code execution via a crafted
.appl file. This provides remote root access on some platforms such as
FreeBSD (used for TrueNAS).
- Fix CVE-2023-42464: Validate data type in dalloc_value_for_key()
https://github.com/Netatalk/netatalk/blob/netatalk-3-1-17/NEWS
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit d170cde027)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
A vulnerability was found in OpenSC. This security flaw cause a buffer
overrun vulnerability in pkcs15 cardos_have_verifyrc_package. The
attacker can supply a smart card package with malformed ASN1 context.
The cardos_have_verifyrc_package function scans the ASN1 buffer for 2
tags, where remaining length is wrongly caculated due to moved starting
pointer. This leads to possible heap-based buffer oob read. In cases
where ASAN is enabled while compiling this causes a crash. Further info
leak or more damage is possible.
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 9c4c3c4c9c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- Fix CVE-2023-40359: xterm before 380 supports ReGIS reporting for
character-set names even if they have unexpected characters (i.e.,
neither alphanumeric nor underscore), aka a pointer/overflow issue.
This can only occur for xterm installations that are configured at
compile time to use a certain experimental feature.
- Update COPYING hash (update in year and version)
https://invisible-island.net/xterm/xterm.log.html#xterm_384
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 164d635f37)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
CVE-2022-1304 only affects e2fsprogs 1.46.5.
The mentioned patch was removed in 6a21733f when bumping to 1.47.0.
Signed-off-by: Daniel Lang <dalang@gmx.at>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit dc0c755273)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
CVE-2021-38185 affects cpio <= 2.13.
The mentioned patches were removed in b0306d94 when bumping to 2.14.
Signed-off-by: Daniel Lang <dalang@gmx.at>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 880e03ba75)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
As of 2021-05-14 CVE-2017-3139 is no longer listed as affecting bind, only RHEL.
Signed-off-by: Daniel Lang <dalang@gmx.at>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 8bf82aab0c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- Fix CVE-2023-36664: Artifex Ghostscript through 10.01.2 mishandles
permission validation for pipe devices (with the %pipe% prefix or the |
pipe character prefix).
- Fix CVE-2023-38559: A buffer overflow flaw was found in
base/gdevdevn.c:1973 in devn_pcx_write_rle() in ghostscript. This
issue may allow a local attacker to cause a denial of service via
outputting a crafted PDF file for a DEVN device with gs.
- Fix CVE-2023-38560: An integer overflow flaw was found in
pcl/pl/plfont.c:418 in pl_glyph_name in ghostscript. This issue may
allow a local attacker to cause a denial of service via transforming a
crafted PCL file to PDF format.
https://ghostscript.readthedocs.io/en/gs10.02.0/News.html
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 93ef6997ae)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since 3341ceb1e5 (package/gdb: zlib is mandatory, not optional), zlib
has become a mandatory dependencies of the gdb package.
However, zlib is only needed for the debugger, gdb itself, while the
server, gdbserver, does not use it.
This means that, when building an SDK to be later reused as an external
toolchain, the zlib headers and libraries are present in the sysroot of
the toolchain, tainting the toolchain and making it unsuitable to be
reused.
As Julien noticed, for example, tcl will try and link with zlib if
available, and at build time it is. But at runtime, it is not, and thus
tclsh fails to run; see 7af8dee3a8 (package/tcl: add mandatory
dependency to zlib)
When we only need to build gdbserver, we still need to configure and
build the whole gdb distribution, which means we call the top-level
configure script; that script has no option to disable the detection
of zlib: it wants to either use a system one, or it will build the
bundled one.
So, when we only build gdbserver, we tell configure to not use a system
zlib. This triggers the build of the bundled one, but it is not linked
with gdbserver so in the end it is not used on the target.
Reported-by: Julien Olivain <ju.o@free.fr>
Reported-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
(cherry picked from commit 8ce33fed49)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
HOST_PROG_SHUTDOWN currently references a file that doesn't exist. Fix
by setting it to /usr/libexec/petitboot/bb-kexec-reboot, which this
package already installs but doesn't use.
Signed-off-by: Reza Arbab <arbab@linux.ibm.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
(cherry picked from commit ab91ddd8a8)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Tcl changed its zlib handling in upstream commit [1]. Before this
commit, the HAVE_ZLIB macro was defined only if a zlib headers/library
was found. After that commit, the HAVE_ZLIB macro is unconditionally
defined. The only change is that: if a working zlib library is found
in the toolchain sysroot, it is used. Otherwise, the package will use
a shipped version in [2]. See also [3] and [4].
This tcl commit is included in Buildroot since commit 7fda943b43
"tcl: bump to version 8.6.1".
In Buildroot, we prefer to not use bundled libraries wherever possible,
so add an unconditional dependency to zlib.
Further notes:
This behavior leads to runtime failures, when the package is compiled
with toolchains including zlib in their sysroot. This is because at
configuration time, the package will detect zlib in the sysroot and
link against it, but the library files won't be installed on target.
This happen to be the case with Bootlin toolchains such as [5], as they
also contaions gdbserver, and since 3341ceb1e5 (package/gdb: zlib is
mandatory, not optional), we also build zlib even if only gdbserver is
built (gdbserver does not use zlib, so that's a bug in our gdb
packaging).
This toolchain also happen to be the one used in basic configurations
of the runtime test infrastructure (this issue was found while
attempting to write a runtime test for tcl).
In such cases, running "tclsh" command fails with error message:
tclsh: error while loading shared libraries: libz.so.1: cannot open shared object file: No such file or directory
libtcl library also miss its dependency.
ldd /usr/lib/libtcl8.6.so
libz.so.1 => not found
libm.so.6 => /lib/libm.so.6 (0xb6dad000)
libc.so.6 => /lib/libc.so.6 (0xb6c65000)
/lib/ld-linux.so.3 (0xb6f6c000)
[1] 6f3dea45ce
[2] https://github.com/tcltk/tcl/tree/core-8-6-13/compat/zlib
[3] https://github.com/tcltk/tcl/blob/core-8-6-13/unix/configure.in#L172
[4] https://github.com/tcltk/tcl/blob/core-8-6-13/unix/Makefile.in#L240
[5] https://toolchains.bootlin.com/downloads/releases/toolchains/armv5-eabi/tarballs/armv5-eabi--glibc--stable-2023.08-1.tar.bz2
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 7af8dee3a8)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix CVE-2023-40225: HAProxy through 2.0.32, 2.1.x and 2.2.x through
2.2.30, 2.3.x and 2.4.x through 2.4.23, 2.5.x and 2.6.x before 2.6.15,
2.7.x before 2.7.10, and 2.8.x before 2.8.2 forwards empty
Content-Length headers, violating RFC 9110 section 8.6. In uncommon
cases, an HTTP/1 server behind HAProxy may interpret the payload as an
extra request.
https://www.mail-archive.com/haproxy@formilux.org/msg43864.html
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 8fc24fbd17)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
A flaw was found in LibRaw. A heap-buffer-overflow in raw2image_ex()
caused by a maliciously crafted file may lead to an application crash.
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit bc4110b073)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fixes the following security issue:
CVE-2023-38039: HTTP headers eat all memory
When curl retrieves an HTTP response, it stores the incoming headers so that
they can be accessed later via the libcurl headers API.
However, curl did not have a limit on the size or quantity of headers it
would accept in a response, allowing a malicious server to stream an endless
series of headers to a client and eventually cause curl to run out of heap
memory.
https://curl.se/docs/CVE-2023-38039.html
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 56b0667406)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Mutt 2.2.12 was released on September 9, 2023. This is a bug-fix
release, fixing two crash bugs. One is possible by viewing a crafted
message header, so upgrading is strongly recommended.
Fix CVE-2023-4874: Null pointer dereference when viewing a specially
crafted email in Mutt >1.5.2 <2.2.12
Fix CVE-2023-4875: Null pointer dereference when composing from a
specially crafted draft message in Mutt >1.5.2 <2.2.12
http://www.mutt.org/
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit e2deaf0467)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Drop patch and disable libseccomp when building statically to avoid the
following build failure raised since commit
29834d8a12:
src/seccomp_notify.c: In function 'seccomp_notify_plugins_load':
src/seccomp_notify.c:136:42: warning: implicit declaration of function 'dlopen'; did you mean 'popen'? [-Wimplicit-function-declaration]
136 | ctx->plugins[s].handle = dlopen(it, RTLD_NOW);
| ^~~~~~
| popen
src/seccomp_notify.c:136:53: error: 'RTLD_NOW' undeclared (first use in this function)
136 | ctx->plugins[s].handle = dlopen(it, RTLD_NOW);
| ^~~~~~~~
Fixes:
- http://autobuild.buildroot.org/results/13d3b46990720bba8621c922b5dce54ab650e96d
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 8144dd1b4c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Does not install systemd unit if nodm or xdm is enabled.
Signed-off-by: Yanghao Cheng <yanghao.cheng@aioi-atg.com>
[yann.morin.1998@free.fr:
- don't use $(or ...)
- slightly reword comment
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 6898ca03a2)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
xdm package also installs a init script that utimately starts X server
Signed-off-by: Yanghao Cheng <yanghao.cheng@aioi-atg.com>
[yann.morin.1998@free.fr:
- don't use $(or ...)
- slightly reword comment
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit f9cd154467)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
tests.package.test_python_pytest.TestPythonPy3Pytest runtime tests fails
now with following error:
======================================================================
FAIL: test_run (tests.package.test_python_pytest.TestPythonPy3Pytest.test_run)
----------------------------------------------------------------------
Traceback (most recent call last):
File "/buildroot/support/testing/tests/package/test_python.py", line 137, in test_run
self.run_sample_scripts()
File "/buildroot/support/testing/tests/package/test_python_pytest.py", line 18, in run_sample_scripts
self.assertRunOk(cmd, timeout=self.timeout)
File "/buildroot/support/testing/infra/basetest.py", line 89, in assertRunOk
self.assertEqual(
AssertionError: 1 != 0 :
Failed to run: python -m pytest sample_python_pytest.py
output was:
Traceback (most recent call last):
File "<frozen runpy>", line 189, in _run_module_as_main
File "<frozen runpy>", line 148, in _get_module_details
File "<frozen runpy>", line 112, in _get_module_details
File "/usr/lib/python3.11/site-packages/pytest/__init__.py", line 5, in <module>
File "/usr/lib/python3.11/site-packages/_pytest/_code/__init__.py", line 2, in <module>
File "/usr/lib/python3.11/site-packages/_pytest/_code/code.py", line 36, in <module>
File "/usr/lib/python3.11/site-packages/pluggy/__init__.py", line 16, in <module>
File "/usr/lib/python3.11/site-packages/pluggy/_manager.py", line 10, in <module>
File "/usr/lib/python3.11/importlib/metadata/__init__.py", line 8, in <module>
File "/usr/lib/python3.11/zipfile.py", line 6, in <module>
ImportError: libz.so.1: cannot open shared object file: No such file or directory
Fix that by adding BR2_PACKAGE_PYTHON3_ZLIB dependency.
Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 0dc1213565)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
libjxl was failing to build for riscv targets, since commit ff7c37e57
"package/libjxl: security bump to version 0.8.1". Build was failing with
output:
/build/libjxl-0.8.1/lib/jxl/enc_xyb.cc: In function 'jxl::Image3F jxl::N_SCALAR::TransformToLinearRGB(const jxl::Image3F&, const jxl::ColorEncoding&, float, const JxlCmsInterface&, jxl::ThreadPool*)':
/build/libjxl-0.8.1/lib/jxl/enc_xyb.cc:223:21: error: variable 'std::atomic<bool> ok' has initializer but incomplete type
223 | std::atomic<bool> ok{true};
| ^~
This build failure was due to a missing <atomic> header inclusion. For
some reason, the build failure was observed only with RISC-V toolchains.
This commit fixes the issue by adding an upstream commit, not yet in a
package release. See [1].
Fixes:
http://autobuild.buildroot.org/results/121/12107bc7aea7afae1d2fb935d31b44eee6ea1501
[1] 22d12d74e7
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit de45aea47b)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
An issue was discovered in lldpd before 1.0.17. By crafting a CDP PDU
packet with specific CDP_TLV_ADDRESSES TLVs, a malicious actor can
remotely force the lldpd daemon to perform an out-of-bounds read on heap
memory. This occurs in cdp_decode in daemon/protocols/cdp.c.
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 3557a7b599)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The quoting around the expansion of ${relative_dir} was indeed incorrect
since it was introduced back in 8fe9894f65 (suport/download: fix git
wrapper with submodules on older git versions): it is in fact already
quoted as part of the whole sed expression.
${GIT} can contain more than one item, but we don't care about splitting
on spaces when we just print it for debug, so we can just quote it
rather than add an exception.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit bcee3ca6d6)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Since commit b7efb43e86 (download/git: try to recover from
utterly-broken repositories), we catch errors through an ERR
trap, so we can try and recover from a broken repository. In
that commit, we switched from using "set -e" to "set -E", so
that trap is inherited in functions, command substitutions,
and subshells.
However, the trap is not defined until we have parsed the
options, created the cache directory, and eventually chdir()ed
into it. Athough improbable, it is possible for the git helper
to fail in any of those steps, and that would not get caught.
Fix that
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit daa341cb9b)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
When we generate the taballs off a local working copy of a VCS tree,
the umask is the one that we enforce in out top-level Makefile.
However, it is possible that a user manually tinkers in said working
copy (e.g. to check an upstream bug fix, or regression). If the user
umask is different from the one Buildroot enfirces, such tinkering
can impact the mode bits of the files, even if their content is not
modified.
When we eventually need to create a tarball from said working copy,
the VCS (e.g. git) will only be interested in checking whether the
content of the files have changed before chcking them out, and will
not look at, and restore/fix the mode bits.
As a consequence, we may create non-reproducible archives.
We fix that by enforcing the mode bits on the files before we create
the tarball: we disable the write and execute bits, and only set the
execute bit if the user execute bit is set.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Vincent Fazio <vfazio@xes-inc.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 768f9f80f6)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
If a patch only removes files, it is ignored. Meaning, that the
registered developer isn't automatically picked up when calling
get-developer.
Fix this by also checking if the line starts with ---, as a patch
removing a file has a line starting with --- with the name of the
removed file and one started with +++ /dev/null.
A set is used to store the changed files, which doesn't allow
duplicates. Therefore normal patches aren't affected by this change.
Signed-off-by: Daniel Lang <dalang@gmx.at>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 7bc5ea80c7)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
go1.20.8 (released 2023-09-06) includes two security fixes to the html/template
package, as well as bug fixes to the compiler, the go command, the runtime, and
the crypto/tls, go/types, net/http, and path/filepath packages.
CVE-2023-39318: html/template: improper handling of HTML-like comments within script contexts
CVE-2023-39319: html/template: improper handling of special tags within script contexts
CVE-2023-39321: crypto/tls: panic when processing post-handshake message on QUIC connections
https://go.dev/doc/devel/release#go1.20.0
Signed-off-by: Christian Stewart <christian@aperture.us>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
/home/autobuild/autobuild/instance-2/output-1/host/mips64-buildroot-linux-uclibc/sysroot/usr/include/openssl/dh.h:195:5: note: previous declaration of 'DH_set0_pqg' was here
# pppd defaults to /etc/ppp/resolv.conf, which not be writable and is
# definitely not useful since the C library only uses
# /etc/resolv.conf. Therefore, we change pppd to use /etc/resolv.conf
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
